BSc-thesis: Developing a business model engineering tool for supporting business model engineering in roll out and market phase - an online investment research company case study
Developing a business model engineering tool forsupporting business model engineering in roll out andmarket phase - an online investment research companycase studyDurk BoersmaTwente UniversityPO Box 2177500AE Enschede, The Netherlandsd.firstname.lastname@example.orgABSTRACTEvery business needs a viable business model. Strangely mostliterature is focused on business model design, whereas there isalmost no attention for validation and implementation of businessmodels. The goal of the research as described in this paper is todevelop a business model engineering tool for supporting businessmodel management as a continuous design, implementation andvalidation cycle. The tool is developed for an online investmentresearch boutique in roll out and market phase. This paperdescribes the research as performed in a case study setting byfocusing on the design and evaluation of the business modelengineering tool.KeywordsBusiness model, action design research, case study, businessmodel engineering tool.1.INTRODUCTIONFor a long time, people and businesses make use of businessmodels. One of the first known business model innovations mightbe in the late nineteenth century. The president of AmericanExpress was on a holiday in Europe, and found it hard to get cash,even though he had good letters of credit . American Expressthen created the travelers check, and from that innovation a robustbusiness model was derived.A more recent definition of a business model comes from theseventies. At that time, business models mostly were used todesign and analyze information, communication and processeswithin an organization. Nowadays, business models are moreabout value networks, but the definition from the seventies stillpartly fits with how we use business models now a days.Although there are a lot of publications about business models,most researchers consider business models as static and describethem mostly qualitatively. Almost none of them consider thatbusiness models constantly (need to) change, and thus need to bemanaged actively, e.g. because of changing market ortechnological environments . With this research we strive forfinding a way to monitor a business model in a more structuredand active way.By developing a tool to manage business models and constantlymanaging the business model, the failure rate of new businessesor technologies may be lowered. This is because the real strengthof an organization may be based on the quality of their businessmodel. Or as Henry Chesbrough said: “A mediocre technologypursued with a great business model may be more valuable than agreat technology exploited via a mediocre business model”. 2.RESEARCH OBJECTIVE AND CASEDESCRIPTIONThe objective of this business model engineering case study withaction design characteristics [4, 5] was to build a business modelengineering tool for an online investment research boutique.An investment research boutique is a company that evaluatesinvestment opportunities related to investing in shares ofcompanies listed on stock markets and sells related analyses totheir clients. The investment research boutique for this researchuses a so called freemium business model . The company offerspeople who subscribed to their mailing list via their website a freeweekly investing column. Paying members also get a monthlyanalysis of three stocks that look interesting from a valueinvesting perspective [7, 8]. The companies behind these stockstypically have a high profitability because of a strong competitiveposition and a sensible business model, good management and abelow average market valuation.The investment research company already has a business model.This business model is developed in the time the company started,with use of the STOF-method, which will be described in section4.2 [2, 16]. Since the company moved from R&D and roll out tothe market phase and is also profitable, its business model can beconsidered as viable. But the company does not have the ability totest its business model in different market scenarios. The aim ofthis research is to develop a business model engineering tool – inthe form of a Microsoft Excel spreadsheet – which can be used formanaging the business model of the investment research company– e.g. by testing, fine tuning and supporting furtherimplementation and development of the business model.Main aim of the engineering tool is to help the founders of the fastgrowing investment research boutique to engineer (monitor, test,adapt and fine tune) their business model in order to discoverstrengths and weaknesses. Also doing this in different scenarios inorder to optimally capitalizing on one of their most importantassets: their mailing list with thousands of investors. Results fromthe engineering process could be used to find areas for businessmodel improvement. Furthermore, the testing tool could be usedto predict sales and profit levels in different market scenarios.
3.RESEARCH APPROACHFor this research the four step Action Design Research (ADR)method from Sein et al  may be an interesting starting pointfrom a research methodology perspective. This method initiallyhas the following four stages:1. Problem formulation (formulating a problem perceived inpractice or anticipated by researchers)2. Building, intervention and evaluation (this phase suggestsbuilding an IT artifact, intervention in the organization andevaluating the artifact)3. Reflection and learning (this is a continuous phase next tophase one and two, ensuring that the research processinvolves more than simply solving a problem)4. Formalization of learning (the outcome of the ADR can beused to answer the initial question. Next to that, once theresearch has an outcome, the outcome should be furtherdeveloped into general solution concepts for a class of fieldproblems)An action design research [4, 5] with iterative cycles is used fordeveloping the business model engineering tool, based onqualitative as well as quantitative analysis. Information and detailsfrom expert interviews, literature studies, expert interviews andquantitative modeling were combined to develop the businessmodel engineering tool. After this, it was refined and improved inthree design cycles with expert interviews. Since there is nooption to take part in the research as a consultant, and thus actionresearch cannot take place, this research is more a design studywith action research characteristics.The method as stated above, thus, is not perfectly suiting, andtherefore needs some adaptation. Since the first step, finding aproblem in practice, is already done, this stage can be skipped.Instead the research starts with analysis of the known problem andthe situation as it is (as-is-situation). Once the as-is-situation isanalyzed, the second and third step from the ADR-method of Seinet al  can be used to develop the engineering tool. The outcomeof the research can be used, in this case study, to fine tune thebusiness model in different scenarios. So, the next steps are used:• Analyze the current, already viable business model: Thealready existing business model needs to be analyzed, theunderlying logic needs to be clear.• Build the engineering tool: Build the business modelengineering tool, using the viable business model as a basis.• Analyze output: The output from the engineering tool can beused for adapting and fine tuning the business model,discovering strengths and weaknesses, and finding newopportunities.After describing the results of a business model literature study inthe next section, the results of this business model engineeringresearch will be described, following the three steps as mentionedabove.4. BUSINESS MODEL LITERATUREREVIEWThe business model concept plays an important role in developinga business model engineering tool. Most business model viewsshow a clear approach on how to build a business model [2, 16].Since the business model concepts show this certain logic forbuilding a business model, it can also be used contrariwise, thus toobtain necessary information for developing an engineering tool.4.1What is a business model?The history of business modeling goes back a long time. Centuriesago, people where not even aware of having a business model. Forexample, a baker bakes bread, sells it, and probably makes somemoney from it. The baker, even he does not know it, has a clearbusiness model. The first time people began thinking aboutbusiness modeling was in the late nineteenth century. Theexample of the president of American Express is already stated inthe introduction. The man developed a new way of getting cashmoney, and obtained a viable business model .Later on in history, people became more interested in businessmodeling, and science became involved. One of the first moremodern definitions of a business model was stated by Timmers: “A business model is an architecture for the product, serviceand information flows, including a description of the variousbusiness actors and their roles, a description of potential benefitsfor the various business actors, and a description of the sourcesand revenues”. At that time, business models were used to showinformation, communication and processes within a company.Then, the main purpose of the business model concept was tobuild information technology systems . More recently, Porter related business models to market structures and howcompanies fit into these structures. Within such a context,business models can be used to make strategic choices .Nowadays, a widely used business model definition within thiscontext is the definition by Rosenbloom and Chesbrough : “Abusiness model is a blueprint for how a network of organizationsco-operates in creating and capturing value from technologicalinnovation”.4.2Components of a business modelIn the early business model literature, most attention has beenpaid to empirically defining business models . More recentliterature has a stronger focus on business model design andseveral business model design components have been proposed.Afuah and Tucci  describe business models as systems that arebuilt from different components, such as value, revenue, sourcesand capabilities. Osterwalder and Pigneur  define fourfundamental business model components: product innovation,customer management, infrastructure management and financialmanagement. These four components are used to group all theirsubcomponentsFor this research, a definition more appropriated for this researchis presented by Bouwman et al [2, 16]. They define a modelalmost similar to Osterwalder and Pigneur . The difference isthat their model has less domains or business model components.Though it has less domains, it covers the same areas as the modelof Osterwalder and Pigneur. The STOF-model, depicted in Figure1, is a systematic approach to design business models and astructured view on business models. The STOF-model uses fourdifferent domains or business model components to describe theunderlying logic of business model designs. Each domain has thegeneration of value for customers as well as the value network asa key point. The business model components are:• Service (describes the components of the service concept theorganization offers, its value proposition and the marketsegments that are targeted)• Technology (description of technical architecture, serviceplatforms, devices, applications)
• Organization (description of actors, roles, interactions,strategies and goals, value activities)• Finance (description of investment sources, cost sources,revenue sources, risk sources, pricing)Since the STOF business model framework gives a systematicapproach in designing a business model, it can also be usedcontrariwise. In this case study, where a viable business model isalready defined, the STOF business model framework can be usedto find and denote the necessary variables for developing abusiness model engineering tool.4.3Business models, static or dynamic?While most business model literature has a static and qualitativenature, early stage business model development and validation ofbusiness model designs is mostly lacking in literature as well aspractice [2, 17]. By changes in market, technology and regulatoryenvironments, most business models change continuously andtherefore are dynamic. A typical example of this continuouslychanging business model is the Finnish company Nokia. Thecompany was founded in 1865, and at that time, of course, did notmake money by producing and selling mobile phones but withmanufacturing paper [2, 16].Thus, as business models are actually dynamic, external factorsshould be taken into account. External factor as rapidtechnological developments or increasing market dynamicsrequire an approach in which the business model can be adaptedto changes. Morris et al describe this as the external ‘fit’ .Sustainable business models, according Morris et al, must beconsistent and the fit between the external factors and theconfiguration of key activities. Bouwman et al make use of threeexternal factors in their dynamic STOF-model, market,technology and regulation.Like the four business model components in the STOF-model caninfluence each other, the same goes for the three external factorsin the dynamic STOF-model. In other words, if somethingchanges in one of the three external factors, it may have impact onthe other two external factors.Next to that, business model design does not necessary need totake place in the implementation or market phase, but can alsotake place in the market offering as well as the R&D phases [17,18]. The dynamic STOF-model as depicted in figure 1 shows thispossibility. Actually, business model design only inimplementation or market phase is very risky and costly . Notmanaging the business model after all, also can be costly .Thus, business models can be considered dynamic, and thereforeneed to be managed actively. In this section a way of developing abusiness model structurally is found in the dynamic STOF-model.For developing a business model engineering tool, the sameframework can be used contrariwise.In the next sections, the three steps from the research approach(section 3) will be used for building a business model engineeringtool. Since the second step, actually building the tool, is mostimportant, the three steps a converted into a six-step approach.This approach is depicted in Figure 5, and contains the followingsteps: analyzing the viable business model, developing a cockpit,transforming cockpit variables into calculations, add scenarios,calculate output for the scenarios and, finally, generate andinterpret the output from the business model engineering tool.5. STEP 1) ANALYZING THE VIABLEBUSINESS MODELAs already stated, the online investment research boutique usedfor this case study, already had a viable business model.Considering Bouwman et al [2, 10], business models are aboutadding value. This value is being added by all kinds of underlyingvariables. Most organizations sell something against some price.Before they can sell their product or service, costs need to bemade, sometimes rough materials need be bought. Both the sellingprice and the costs need to be made to sell, are examples ofvariables that can lead to addition in value.First step in developing a business model engineering tool is tofind the most important variables. In this case study most of thevariables are found through expert interviews with the founders ofthe online investment research boutique. The STOF-model fromBouwman et al  helps as a kind of checklist and to group thevariables. In this case, the variables are divided into to the fourcomponents of the STOF-model, with ‘market’ as an extracomponent. An overview can be found in table 1.The variables under the business model component service, aremostly related to the services the online investment researchboutique offers. VP, VS and TopX are abbreviations of productsthey offer. Furthermore their most important asset, the mailinglist,and the growth are considered to be a service component.The variables belonging to the technology component, are coststhat come from technology the organization uses. Variables usedhere are iDeal and Paypal (for receiving payments fromcustomers), costs for webhosting and backup (to keep theirwebsite in the air) and costs of inventory.The organization component contains the costs for keeping theorganization up and running. Salaries, traveling costs,subscriptions on investment magazines and websites as well asmarketing belong to this business model component.The financial component is the most important in this case study,because it contains the most variables and is crucial for theexistence of the organization. This component contains all pricingdetails for the services delivered, as well as all other variables thathave to do with cash in- and out-flows.Once the variables are found, step 2 can be started.6.STEP 2) DEVELOPING A COCKPITThe variables in the first step need to be processed into thebusiness model engineering tool. For this research, theengineering tool will be built in Microsoft Office Excel.According to Tennent and Friend , a good way for showingthe most important variables, is to use a sheet exclusively forthese variables. This sheet will be called the cockpit. The variablescan be shown in exactly the same way as they are found, using thebusiness model components from the dynamic STOF-model.Since the main aim of this business model engineering tool is tomonitor, test, adapt and fine tune the business model, a calculationover several years has to be made. Therefore, a starting year canbe added in the cockpit. Figure 4 depicts the components of thebusiness model.In this business case, the founders of the online investmentresearch boutique wanted to test their business model in differentscenarios. Therefore, the cockpit at this time only contains the
standard variables as found in the first step. The standardvariables, are the only number that can be changed in the cockpit.Last thing that has been done in this step, is an expert interviewwith one of the founders of the online investment researchboutique. By this interview, the variables were checked, andcockpit checked for completeness.7.STEP 3) CALCULATIONSWith the cockpit and the variables the first part of the businessmodel engineering tool is completed. But, since the businessmodel engineering tool needs to be used for monitoring, testing,adapting and fine tuning the business model, there has to be someoutput. By following the principles of Tennent and Friend, a newsheet is created for output variables.Most of the outputs can be created by using one or more variablesfrom the cockpit, and adding a formula. For example, using thenumber of subscriptions for a certain service, the price people payfor their subscription, and multiplying these two, the turnover forthat service is known. For the structure of the engineering tool, thesame business modeling components as in the first and secondstep can be used again.Furthermore, the most logical way is to begin with all things thatcreate a positive cashflow. Therefore, in the case study, all of theservices the investment research boutique offers, are calculated.These turnover add up to a total turnover. Then the costs can besubtracted from the total turnover. Referring to the dynamicSTOF-model, the total turnover in this case was subtracted bytechnology cost, organization cost and financial components.In accounting terms, when you subtract all the costs from the totalturnover, the gross margin is what is left. This gross margin needsto be subtracted by the value added tax (VAT) and tax on income,which leaves the profit after tax. In the case of the onlineinvestment research boutique, the profit after tax is divided by thesize of the mailing list, to calculate a net margin per emailaddress.8.STEP 4) ADDING SCENARIOSOnce the business model engineering tool is up and running, thetool can be made more dynamically by adding scenarios.According to Tennent and Friend , one of the most functionalways to create scenarios is by putting two main variables in amatrix. In a way of finding two good variables for this businesscase, a story from one of the founders of the investment researchboutique gave inspiration.‘An investor always looks for great opportunities to invest hismoney. One of these investors always addressed the samequestion before making an investment decision. The question was:which two possible developments would affect the organization,but are completely out of your control’. These developmentscould, for example, be a change in regulation by the government.In this case study, people need to have money before they willsubscribe to some service. Thus, the economic growth must bepositive. Next to that, people tend to invest more if the stockmarket is doing well. Considering these two findings, both marketsentiment and economic growth tend to be important, while bothof them cannot be controlled by the investment research boutique.Therefore, the variables market sentiment and economic growthwere most usable. When these two variables are put in a matrix,the situation as depicted in Figure 2 will appear.• Scenario 1: In this scenario the market sentiment is positive,while the economic growth is negative. A situation as thishappened in the second quarter of 2009. Stock indices werealready growing and thus the market sentiment was positive,while because of the crisis the economic growth was negative.• Scenario 2: In this scenario the market sentiment as well as theeconomic growth are positive. A situation like this scenario canbe found in the internet bubble. Stock indices raise sky-high,market sentiment was very good and lots of people made quitesome money during this period (after losing it again later).• Scenario 3: This is the worst scenario, because both marketsentiment and economic growth are negative. This situation islike the stock market crash late 2008. Stock indices crashed,banks went bankrupt en people had absolutely no trust in thefinancial world.• Scenario 4: In this scenario the economic growth is positive,while the market sentiment is negative. A comparison to thissituation can be found in the last half of 2002 and the first halfof 2003. People still were in the stock market, while most stockindices were already lowering.Once these scenarios are denoted, the variables in the businessmodel engineering tool will change in each scenario. In thebusiness model engineering tool for the online investmentresearch boutique, variables can be entered into the model asstandard variables. The model then can be set to a certainscenario, and the standard variables will be changed by acombination of a market sentiment and economic growthmultiplier.When the economic growth is negative, people tend to spend lessmoney and probably less people will have subscriptions to aservice of the online investment research boutique. Lesssubscriptions means less payments, and therefore the existence ofthe organization could be threatened. On the other hand, when themarket sentiment is positive and the economic growth is alsopositive, more people tend to invest their money. In this scenario,probably more subscriptions will be sold, and therefore theincoming payments will rise.As the market sentiment and economic growth are set, thecalculation begins. In an overview of output, a completecalculation of turnover, profit, taxes and profit after tax are shown.By changing the scenario in the cockpit, an outcome can begenerated for every scenario. Once the output per scenario isknown, suggestions for fine tuning and improving the businessmodel can be made.9. STEP 5) SCENARIO-BASEDCALCULATIONIn the cockpit, the four scenarios as stated in the previous section,lead to a switch for choosing the scenario. Furthermore, moststandard variables have are not correct for each scenario.Therefore they need to be corrected. To make the engineering toolmore dynamic, this correction should also take into account themarket sentiment and the economic growth.Therefore a new table was entered into the cockpit, as depicted inFigure 3. For each of the four scenarios as depicted in Figure 2, amultiplying factor can be entered into the cockpit. To make itmore realistic, for both market sentiment and economic growththe weight can be set. The result of this table will be a generalmultiplier that can be used for transforming the standard variablesinto scenario-based variables. Most of the standard variables aremultiplied by the general multiplier belonging to the scenario, toget a scenario-based variable.
For some of the standard variables only one of the two scenario-factors will influence them. Therefore it is also possible to onlymake use of market sentiment multiplier or economic growthmultiplier.10.STEP 6) GENERATING OUTPUTSince the variables are known, dynamic elements are included,scenario-based calculation is used, the business model engineeringtool is almost finished. An import aspect that still has to be donein this step is to generate output which can be used to monitor,test, adapt and fine tune the business model.Analogous to creating input and calculation sheets, the output isprojected on a separate sheet. For monitoring and testing thebusiness model, it would be most useful to create some charts thatshow important output, growth or decrease. In this business casethe key development is the margin per email address.For adapting and fine tuning the business model, calculations inonly one scenario is not enough. Therefore in the onlineinvestment research boutique case, four extra calculation sheetsare developed. Therefore each scenario can be calculated andafterwards compared with the other scenarios. This information isuseful to discover strength and weaknesses in the business model,and thus useful for adapting and fine tuning the business model. Inthe next section, calculations from the business case will becompared, and tips for fine tuning or adaptation will be given.11.SCENARIO ANALYSISTo adapt and fine tune the business model so that it fits everyscenario, can only be done if the output from all of the scenarios iscompared. In this section the strength and weakness of eachscenario will be denoted, the scenarios will be compared to eachother, and tips for adaptation and fine tuning will be given. Thecalculations within each scenario have been done by a fundedguess about the market sentiment and economic growthmultipliers. This is not exactly correct, but as the great WarrenBuffet once said:“It is better to be approximately right than precisely wrong.”11.1Output from scenario 1Since scenario one only happens with negative economic growthand positive market sentiment, people will still invest, but only ifthey have the money. In the cockpit, this means that the growth ofthe mailing list and number of subscriptions is low, but there is agrowth.When taking a look in the calculations, because of the smallgrowth, the most important variable (margin per email address) isgrowing. Over a period of five years, the net margin per emailaddress has grown by almost 52 percent. Next to that, the profitafter tax has grown by almost 65 percent, as depicted Chart 1.Chart 1: Profit after tax, scenario 1If taken into consideration that the profit after tax is actuallymoney that is not needed for some purpose, one possibility is toinvest that money into a ‘management portfolio’. This portfoliocan also be used when money is needed in worse times, as somekind of buffer. When doing this, the yearly addition as shown inChart 2 will take place, which results in a potential buffer of overtwo million euros after five years.Chart 2: Additions on MNGT, scenario 1In this scenario, there are almost no weaknesses. Money is made,key figures are growing and eventually profit will be made. Whenmonitoring the business model in this scenario, a greatopportunity lies in making the mailing list grow. When themailing list grows, the number of subscriptions will increase andafter all, the profit and margin per email address will increase.11.2Output from scenario 2From all of the four scenarios, this is the most positive. Sincemost business models are viable in this scenario, same counts forthis business model. Still it can use a little explanation. Bothmarket sentiment and economic growth are positive, ensuring thatmore people will start investing.When looking into the calculations belonging to this scenario, noweaknesses can be found. Every variable is growing and positive.The key variable, margin per email address, will grow with 242percent over a period of five years. The profit after tax will evengrow with 285 percent in five years, as depicted in Chart 3.Chart 3: Profit after tax, scenario 2The chart clearly shows a more exponentially growing line, whichis positive for the existence of the investment research boutique.When looking to the possible additions on the managementportfolio, as depicted in Chart 4, the same exponential line isshown. In five years time, the management portfolio in scenariotwo, could grow to almost five million euros.
Chart 4: Additions on MNGT, scenario 2Since this is the best possible scenario, weaknesses are not found.When the investment boutique is in this scenario, the only goodthing to do is keeping the mailing list growing. Even when doingnothing at all, profits will still be growing. But since almost everybusiness model can be profitable in this kind of scenario, this isnot the most interesting scenario.11.3Output from scenario 3This scenario is the direct opposite of the previous scenario. Boththe market sentiment and the economic growth are negative.When market sentiment is negative, less people tend to startinvesting. Next to that, because of negative economic growth, lesspeople do not have the money to start investing.The calculations belonging to scenario three, give a clear view onthe miserable situation. Growth of the mailing list is minimal,number of subscriptions compared to standard is much lower,profit drops and thus the margin per email address drops as well.This margin could drop till around ten cents per address, adecrease of almost 96 percent. Profit after tax could also decreasewith almost 96 percent, as shown in Chart 5, resulting in a profitafter tax from around 3000 euros a year.Chart 5: Profit after tax, scenario 3When taking a look at the management portfolio which can beseen as a sort of money buffer, scenario three is the situation todiscover that this portfolio is a must have. The addition to theportfolio through profit, at first it is still positive, but in later yearsturns negative. Next to that, the return on the portfolio is alsonegative. Negative addition means that money is subtracted fromthe portfolio to put in back in the investment research boutique.When this portfolio had not been made in front, the businesswould probably be bankrupt! The additions and subtractionsaccording the management portfolio are shown in Chart 6.Chart 6: Additions on MNGT, scenario 3When looking at the calculations in scenario three, one thing isclear. Without the management portfolio the investment researchboutique would be bankrupt in this scenario. Thus, in betterscenarios, for example one or two, the boutique can invest profitinto the management portfolio to grant the coexistence duringrougher scenarios.Furthermore, the business model engineering tool showssomething else that needs to be discussed. First, because of thelow costs and the management portfolio, the boutique can surviverough scenarios. But, when more costs are cut, the chance ofsurviving is even bigger. The calculations show that there still aresome costs that can be lowered, like the traveling and subscriptioncost. Second, making the list grow would probably generate moreprofit. Even in bad economic situations, or maybe just because ofthe bad economic situations, real investors see new opportunities.These people need to be attracted to the boutique.11.4Output from scenario 4The last scenario given by the matrix depicted in Figure 2, has anegative market sentiment and a positive economic growth. Inother words, people have some money they can invest, but thestock indices are not that good. When looking at the calculationsfor this scenario, numbers are not that good, but the existence ofthe boutique is not immediately at risk.From the calculations, the conclusion can be made that the growthof both profit after tax and margin per email address is negative.Margin per email address dropped little over 48 percent, whileprofit after tax dropped little over 45 percent. The decrease ofprofit after tax is depicted in Chart 7.Chart 7: Profit after tax, scenario 4When the profit after tax is decreasing, at some point in time,profit will be too low to cover the costs. That is why the earlierdenoted management portfolio is a good buffer. When includingthe variables from scenario four in the calculation for additions tothis management portfolio, shows that the portfolio is stillgrowing. Additions through return on investment are not much,
but still the line is slightly climbing. The addition through profit,since the profit is slightly decreasing, is also decreasing. This isdepicted in Chart 8.Chart 8: Additions on MNGT, scenario 4From the business model engineering tool can be concluded thatthe business model can still survive, even in a slightly badscenario as this one. Main reason for this survival is the low costsof operating the investment research boutique. More profit couldbe gained in this scenario, again, by increasing the size of themailing list.11.5Comparing four scenariosIn a short recapitulation, the business model tested in this casestudy is profitable in all scenarios, during a period of five years.The profitability in scenario three is only because of the existingmanagement portfolio, without this portfolio the onlineinvestment research boutique would probably be bankrupt.When comparing the four scenarios and their outputs to eachother, a number of interesting things can be said about thisbusiness model. First of all, because of the relatively low enconstant costs, the profitability of the business model is mostlydependent on the size of the mailing list and the number of peoplewith a subscription. Although the costs are relatively low, in reallybad times there are some possibilities to save money on certaincosts. Second, according to the business model engineering tool itseems a great idea to invest the profit after tax in a certain buffer.A management portfolio as suggested in an earlier section couldbe an option, but is relatively uncertain with respect to stockcrashes.Considering the most important output variable, the margin peremail address, different scenarios can have an immense influenceon that. Chart 9 gives the expected growth and increase perscenario in the next five years. Over a period of five years, themargin could vary somewhere from around ten cents to thirtyeuros.Chart 9: Margin per email address per scenarioAnother interesting view on the output, is the one on the value ofthe management portfolio. Although it initially could be used as abuffer, in good times it can become an enormous valuable asset.The value in five years from now, could vary somewhere between160,000 euros and 5 million euros.12. CONCLUSIONSIn this paper, an approach for developing a business modelengineering tool is given. With this engineering tool,organizations can monitor, test, adapt and fine tune their businessmodel. The first results are encouraging: the case study shows thatthe business model can be tested in different scenarios, andstrengths and weaknesses were discovered. Based on the analysisthe business model can be further improved. Therefore it issupposed that engineering a business model with this tool maylead to a more viable business model and thus a healthierbusiness.In the first step of the six-steps framework, as depicted in Figure5, the given business model is analyzed. This analysis needs to bedone before the main variables can be stated in step two. To statethis variables, a clear view of how business works needs to beobtained. The analysis and stating the variables is a really timeexpensive step. In the case study, expert interviews with one of thefounders are used, even in a cyclic character. Since this case studyconsiders a relatively small organization, this step would eventake more time when a multinational is involved.Once the variables are stated, you can directly transfer them to theengineering tool. In the six-step model, these variables weretransferred in the same logical way as they were obtained from thebusiness model. Therefore the dynamic STOF-model fromBouwman et al  was used. The third step is to makecalculations with the standard variables from the previous step.This step is relatively easy and quick, if the first two steps aredone correctly.Once the calculations are made, the business model engineeringtool is up and running. To obtain usable information formonitoring, testing, adapting and fine tuning the business model,some kind of uncertainty has to be added. In this business case,this uncertainty comes from four different scenarios. Thesescenarios are created in step four. As these scenarios cannot beexactly described since the future is uncertain by nature, it isimportant to take two key variables. In the business case, thesevariables are obtained by searching for two factors that have greatinfluence on the investment research boutique, but cannot beinfluenced by the boutique. Once these two factors are stated, amultiplier needs to be developed. In this business case, themultiplier is a weighted average from the two factors. This step isnot extremely difficult, but takes relatively much time because theforecasting needs to be done pretty realistically.The fifth step is to make the calculations from step three, alsopossible with the scenario-based variables. This is not difficult,because the calculations are already made, only thing to do isformalize them for all variables. Once all calculations are made,the last step can be done through creating useful output. Next tomaking output per scenario, another useful view is the output ofall four scenarios in one chart or table.Although the business model engineering tool gave newinformation, strengths and weaknesses about the business model,there is a disadvantage. Obtaining variables from the viablebusiness model, making scenarios and scenario-based calculationsis pretty time consuming– even though this case study contains arelatively small organization. Thus, developing a business model
engineering tool for a large organization takes much more time.Since engineering a business model is not a common thing, andnew valuable information can be obtained from it, some kind ofbalance needs to be found between the efforts and the valuableinformation.13. REFERENCES J. Margretta, Why Business Models Matter, HarvardBusiness School Publishing Corporation, 2002. B. Kijl, H. Bouwman, T. Haaker and E. Faber, Developinga dynamic business model framework for emergingmobile services, ITS 16th European Regional Conference,Porto, Portugal, 2005. H. Chesbrough, Business Model Innovation: Oppertunitiesand Barriers, Long Range Planning (2009), doi:10.1016/j.lrp.2009.07.010 R. Cole, S. Purao, M. Rossi and M. Sein, Being proactive:where action research meets design research, ICIS 2005,Las Vegas, USA, 2005, pp. 325-336. M. Sein, O. Henfridsson, S. Purao, M. Rossi, R.Lindgren, Action Design Research. Paper under review forMISQ Magazine. C. Anderson (2009), Free: The Future of a Radical Price.New York, Hyperion Books. H. Oude Nijhuis, B. Kijl (2009). Warren Buffett: Zijnbeleggingsstrategie in theorie & praktijk. Arnhem,Sonsbeek Publishers. B. Graham (1949). The intelligent investor. (2003 edition)United Kingdom, HarperCollins. Timmers, P. (1998). Business models for E-commerce. InElectronic Markets, 8(2). pp 3-7 Bouwman, H., Vos, H. de & Haaker, T. (2008). Businessmodel concept, typologies and components. In Mobileservice innovation and Business Models. BerlinHeidelberg: Springer. Porter, M. (2001). Strategy and the internet. In HarvardBusiness review, March. pp 63-76 H. Chesbrough and R.S. Rosenbloom, “The role of thebusiness model in capturing value from innovation:evidence from Xerox Corporation’s technology spin-offcompanies”. Industrial and Corporate Change, 11 (2002),pp. 529-555 J. Hedman and T. Kalling, “The business model concept:theoretical underpinnings and empirical illustrations”,European Journal of Information Systems, 12 (2003), pp.49-59 A. Afuah and C. Tucci, Internet business models andstrategies: Text and cases, McGraw-Hill HigherEducation, 2000. A. Osterwalder and Y. Pigneur, An e-business modelontology for modeling e-business, 15th Bled ElectronicCommerce Conference, Bled, Slovenia, 2002, pp. 17-19 H. Bouwman, E. Faber, T. Haaker, B. Kijl and M. deReuver, Conceptualizing the STOF model, in H.Bouwman, T. Haaker and H. de Vos, eds., Mobile ServiceInnovation and Business Models, Springer Verlag, Berlin,2008, pp 31-70. H. Mason and T. Rohner, The venture imperative: a newmodel for corporate innovation, Harvard Business SchoolPress, 2002. Tennent, J., Friend, G. (2005). Guide to businessmodelling. Second edition. London: Profile Books LTD. Morris, M., Schinderhutte, M., and Allen, J., Theentrepreneur’s business model: toward a unifiedperspective. Journal of Business Research, 2005. 58(6):pp. 726-735
Figure 1: The dynamic STOF-model from Bouwman et al [1, 15]Figure 2: Scenario building through variables in a matrixFigure 3: Multiplying factor calculation
Figure 4: Part of the cockpit of the business model engineering tool
Figure 5: 6 step business model engineering tool frameworkService Technology Organization Financial MarketSize mailinglist Inventary Number of FTE VS Price Y Market sentimentVS subscr Y Webhosting Salary VS Price Q No. investorsVS subscr Q iDeal Traveling cost VS Price M Eco. growthVS subscr M Paypal Subscriptions VP Price YVP subscr Y Marketing VP Price QVP subscr Q VP Price MVP subscr M FATTopX subscr Tax on incomeTable 1: Variables derived from the business model of the online investment research boutique